CHAPTER 5.

0 DESIGN OF SINGLE PILES

5.1 INTRODUCTION The methods to determine the static capacity of single piles presented in this chapter have been selected because of their simplicity and excellent track record for predicting pile capacity when compared to pile load tests. A step-by-step procedure is presented for each method. Each procedure is taken from the FHWA manual “Design and Construction of Driven Pile Foundations (FHWA-HI-97-013). The methods presented in this manual and when they are applicable is provided in Table 5-1. Table 5-1 Design Methods Cohesionless Soil Cohesive Soil Yes Yes No Yes Yes No No Yes Yes Yes

Design Method Meyerhof Nordlund α Method Effective Stress Nottingham & Schmertmann

Applicable for Final Design No Yes Yes Yes Yes

5.2 MEYERHOF METHOD FOR PILES IN COHESIONLESS SOILS (Meyerhof, 1976) Meyerhof developed a method of estimating pile capacity based on empirical correlations between standard penetration test (SPT) results and static pile load tests. The advantages of this method are that it is very easy to use and that SPT data is typically available for a project. The major disadvantage of this method is that SPT values are non-reproducible and can be influenced by many factors (i.e., rod length, hammer efficiency, overburden depth, etc.). Because of the simplicity of the method, many simplifying assumptions are contained in the method, resulting in a less reliable method than the other methods presented in this manual. This method should be used for preliminary estimates and not for final design. For displacement piles (e.g., timber piles) Meyerhof has established that the average unit shaft resistance (fs) is: fs = 2 N' 50 ≤ 2 ksf (5-1)

N' is the average corrected SPT resistance in blows per foot
The unit toe resistance (qt) in ksf for piles driven into sands and gravels may be approximated by the following equation:

31

qt = 8 N' o + [ (0. 8 N' B −0. 8 N' O ) D B ] ÷ b ≤ 8 N' B where:

(5-2)

N' O = Average corrected SPT N’ value for the stratum, overlying the bearing stratum N' B = Average corrected SPT N’ value for the bearing stratum D B = Pile embedment depth into the bearing stratum in feet b = Pile tip diameter in feet
Equation 5-2 applies when the pile toe is located near the interface of two strata, with the weaker stratum above the bearing stratum. The limiting value of the unit toe resistance is reached when the embedment depth into the bearing stratum reaches 10 pile diameters. For piles driven into a uniform cohesionless stratum, the unit toe resistance in ksf is determined from the following equation: qt = 0. 8 N ' B D B ÷ b ≤ 8 N ' B (5-3)

It is recommended that the average corrected SPT N’ value N' B be calculated by averaging N’ values within the zone extending 3 diameters below the pile toe. For piles driven into non-plastic silts, Meyerhof recommended the unit toe resistance, qt, be limited to 300 N' B instead of the 400 N' B given in the above equation. STEP BY STEP PROCEDURE FOR MEYERHOF METHOD (FHWA-HI-97-013) Step 1 Correct SPT field N values for overburden pressure. Use correction factors from Figure 5-1. N’ = CN N where: N’ CN N = corrected SPT N value = correction factor for overburden stress (Figure 5-1) = uncorrected or field SPT value

Step 2

Compute the average corrected SPT N’ value (N' ) for each soil layer. Along the embedment length of pile delineate the soil profile into layers based on density indicated by the N’ value. The individual soil layers should be selected between 10 and 20 feet. Compute the unit shaft resistance in ksf for timber piles from: fs = 2 N ′ 50 ≤ 2

Step 3

Step 4

Compute ultimate shaft resistance Rs (kips) Rs = fs As

32

where: Step 5

As

= pile shaft surface area = (perimeter) x (embedded length)

Compute the average corrected SPT N’ values (N' O ) and (N' B ) near the pile toe. In cases where the pile toe is situated near the interface of a weaker stratum overlying the bearing stratum, compute the average corrected SPT N’ value for the stratum overlying the bearing stratum, (N' O ) , and the average corrected SPT

N’ value for the bearing stratum (N' B ) .

In uniform cohesionless soils, compute the average corrected SPT N’ value by averaging N’ values within the zone extending 3 diameters below the pile toe. Step 6 Compute the unit toe resistance qt For weaker soils overlying the bearing stratum, compute qt from:

′ qt = 8 N' o +(0. 8 N' B −0. 8 N' O ) D B ÷ b ≤ 8 N B
For pile in a uniform cohesionless deposit, compute qt from:

′ qt = 0. 8 N ' B D B ÷ b ≤ 8 N B
For pile driven into non-plastic silts, the unit toe resistance should be limited to 0. 6 N' B ksf. Step 7 Compute the ultimate toe resistance Rt (kips) Rt = qt At Where At is the pile toe area (ft2). Step 8 Compute the ultimate pile capacity (kips) Qu = Rs + Rt Step 9 Compute the allowable design load Qa (kips) Qa = Qu / factor of Safety The Meyerhof Method should be used only for preliminary capacity and length estimates.

33

Figure 5-1: Chart for Correction of N-Values in Sand for Influence of Overburden Pressure (from Peck, Hanson, Thornburn, 1974) 5.3 NORDLUND METHOD FOR PILES IN COHESIONLESS SOILS (1963) The Nordlund method considers the type of the pile (i.e. coefficient of friction between the pile material and soil, displacement versus non-displacement, etc.) and the soil pile interaction in calculating the shaft resistance. The shaft resistance of a pile is a function of several parameters including the following:

• • • • • • •

Friction angle of the soil Friction angle of sliding surface (soil/pile interface) Taper of the pile Effective unit weight of the soil Pile length Minimum pile perimeter Volume of soil displaced

The Nordlund method attempts to take these parameters into consideration when evaluating pile capacity. This method is a semi-empirical approach that is widely used. The Nordlund Method (Figure 5-2) equation for computing the ultimate capacity of a pile is as follows:
D Q u = [ ∑ d =0 K δ C F p d sin (δ + ω )C d ∆d ÷ cos ω ] + α t N' q At p t d=

(5-4)

where: d D Kδ = = = Depth Embedment pile length Coefficient of lateral earth pressure at depth d

34

CF pd δ ω ϕ Cd ∆d αt N’q At pt

= = = = = = = = = = =

Correction factor for Kδ when δ ≠ ϕ Effective overburden pressure at the center of depth increment d Friction angle between pile and soil Angle of pile taper from vertical Soil effective friction angle Pile perimeter at depth d Length of pile segment Dimensionless factor (dependent on pile depth-width relationships) Bearing capacity factor Pile toe area Effective overburden pressure at the pile toe (limited to 3 ksf).

Figure 5-2 Nordlund’s General Equation for Ultimate Pile Capacity (FHWA-HI-97-013) STEP BY STEP PROCEDURE FOR NORDLUND METHOD (FHWA-HI-97-013) Step 1 Delineate the soil profile into layers and determine the ϕ angle for each layer. A.) Construct an effective overburden pressure (po) diagram versus depth. B.) Correct the SPT field N values for overburden pressure using Figure 5-1. Delineate soil profile into layers based on corrected SPT N‘ values

35

C.) Determine ϕ angle for each layer of soil from laboratory tests or in-situ data. D.) In the absence of laboratory or in-situ test data, determine the average corrected SPT N’ values (N' ) for each soil layer and determine ϕ angle from Table 5-1. Table 5-1 Empirical Correlation for Effective Friction Angle of Granular soils based on Corrected SPT Value (after Bowles, 1977) Description Corrected N’ ϕ angle** Very Loose 0 to 4 25 - 30° Loose 4 to 10 27 - 32° Medium 10 to 30 30 - 35° Dense 30 to 50 35 to 40° Very Dense 50+ 38 - 43°

* Corrections may be unreliable in soils containing gravel. ** Use larger values for granular material with 5% or less fine sand and silt.

Step 2

Determine the friction angle between the pile and soil (δ) based on the displaced soil volume (V) and the soil friction angle (ϕ). A.) Compute the volume of soil displaced per unit length of pile (V). B.) Use Figure 5-3 to determine the ratio of the pile soil friction angle to the soil friction angle δ/ϕ. C.) Calculate δ based on δ/ϕ ratio.

Step 3

Determine the coefficient of lateral earth pressure Kδ for each ϕ angle. Determine Kδ for each ϕ angle based on displaced volume ,V, and pile taper angle (ω) using Figures 5-4 – 5-7 and the appropriate procedure in steps 3 A, B, or C. A.) If the displaced volume is 0.1, 1, or 10 ft3/ft, which corresponds to one of the curves in Figures 5-4 through 5-7, and the soil friction angle is one of those provided, Kδ may be determined directly from the appropriate figure. B.) If the displaced volume is 0.1, 1, or 10 ft3/ft, which corresponds to one of the curves provided in Figures 5-4 through 5-7, but the effective friction angle (ϕ) is different from those provided, use a linear interpolation to determine Kδ for the required ϕ (see FHWA-HI-97-013 for additional detail). C.) If the displaced volume is other than 0.1, 1, or 10 ft3/ft, which corresponds to one of the curves provided in Figures 5-4 through 5-7, but the effective friction angle (ϕ) is one of those provided, use a log linear interpolation to determine Kδ for the required volume (see FHWA-HI-97-013 for additional detail). For preliminary designs Kδ may be estimated by visual estimation between curves in Figures 5-4 through 5-7.

Step 4

Determine the correction factor, CF, to be applied to Kδ if δ ≠ ϕ. Use Figure 5-8 to determine the correction factor for each Kδ.

36

Step 5

Compute the average effective overburden pressure at the mid-point of each layer (pd). Compute the shaft resistance in each layer of soil. The sum of the shaft resistance from each layer obtained is equivalent to the ultimate shaft resistance Rs.
D R s = ∑d =0 K δ C F p d sin (δ + ω )C d ∆d ÷ cos ω d=

Step 6

Step 7

Determine the αt coefficient and the bearing capacity factor, N’q, from the friction angle of the soil near the pile toe. A.) Use Figure 5-9a to determine αt coefficient based on pile length to diameter ratio. B.) Use Figure 5-9b to determine N’q. C.) If the friction angle of the soil is estimated from SPT data, compute the average corrected SPT N’ value over the zone from the pile toe to 3 diameters below the pile toe. Use this average corrected N’ value to estimate the friction angle near the toe of the pile using Table 5-1

2.50

2.00

VOLUME, V (ft3/ft)

1.50

1.00

0.50

0.00 0.00

0.25

0.50

0.75

1.00

1.25

1.50

δ /φ

Figure 5-3: Relationship of δ/ϕ and pile displacement (V) for timber piles (after Nordlund, 1979)

37

7 6 5

4 3 2

1.00 0.85 1 0.70 0 0.0 0.5

V = 10.0 ft3/ft V = 1.0 ft3/ft V = 0.1 ft3/ft

ω (degrees)

1.0

1.5

2.0

Figure 5-4: Design curve for evaluating Kδ for piles when ϕ = 25°(after Nordlund, 1979)

7 6 5

4 3

2 1.45 1.15 0.85 1 0 0.0

V = 10.0 ft3/ft V = 1.0 ft3/ft V = 0.1 ft3/ft 0.5 1.0 1.5 2.0

ω (degrees)

Figure 5-5: Design curve for evaluating Kδ for piles when ϕ = 30°(after Nordlund, 1979)

38

14 12 10


2.35 1.75 1.15

8 6 4 2 0 0.0

V = 10.0 ft3/ft V = 1.0 ft3/ft 3 V = 0.1 ft /ft 0.5 1.0 1.5 2.0

ω (degrees)

Figure 5-6: Design curve for evaluating Kδ for piles when ϕ = 35°(after Nordlund, 1979)

20

15


10

4.30 3.00 1.70

5

V = 10.0 ft3/ft V = 1.0 ft3/ft 3 V = 0.1 ft /ft

0 0.0

0.5

ω (degrees)

1.0

1.5

2.0

Figure 5-7: Design curve for evaluating Kδ for piles when ϕ = 40°(after Nordlund, 1979)

39

Step 8

Compute the effective overburden pressure at the pile toe pt. Note that the limiting value of pt is 3 ksf.

Step 9

Compute the ultimate toe resistance Rt with the following two steps: A.) B.)

R t = α t N' q At p t
Limit R t = q l At where q l is obtained from Figure 5-10 and the following two steps: 1.) With the friction angle near the toe of the pile determined from laboratory or in-situ test data; and 2.) With the friction angle of the soil estimated from SPT corrected values (N’) and Table 5-1. Use the lesser of the two values.

Step 10

Compute the ultimate pile capacity (kips) Qu = Rs + Rt

Step 11

Compute the allowable design load Qa (kips) Qa = Qu / Factor of Safety

Figure 5-8: Correction Factor for Kδ when δ ≠ ϕ (after Nordlund, 1979)

40

5.4

α- METHOD FOR PILES IN COHESIVE SOILS (Tomlinson, 1979)

The ultimate bearing capacity of a pile in cohesive soil may develop up to 80 – 90% of its capacity through shaft resistance. The α-Method is a total stress analysis where the ultimate capacity of the pile is determined from the undrained shear strength of the cohesive soil. This method assumes that the shaft resistance is independent of the effective overburden pressure. The unit shaft resistance is expressed in terms of an empirical adhesion factor times the undrained shear strength. The unit shaft resistance, fs, is equal to the adhesion (ca) which is the shear stress between the pile and the soil.

f s = c a = αc u

(5-6)

α is an empirical adhesion factor to reduce the average undrained shear strength (cu) of the undisturbed clay along the embedded length of the pile. The coefficient α depends on the nature and strength of the clay, pile dimensions, method of installation, and time effects. Figure 5-11 should be used to determine the pile adhesion for the general case of a homogeneous soil profile. Figure 5-12a should be used when driving a pile through a layer of sand or sandy gravel which is above a stiff clay layer. This condition will typically develop the highest adhesion factors as the granular soil is dragged into the underlying clay. Figure 5-12b should be used for determining the adhesion for piles driven through soft clay into stiff clay. In this case, the soft clay is dragged into the stiff clay stratum reducing the adhesion factor of the underlying stiff clay.
Figure 5-12c may be used when driving piles in stiff clays without any different overlying strata. In stiff clays, a gap often forms between the pile and the soil in the upper portion of the pile. The adhesion factor is, therefore, reduced at shallow pile penetration depths and increased at deeper pile penetration depths. The unit toe resistance is determined for homogeneous cohesive soil using the following equation: q = cu N c (5-7) The term Nc is a dimensionless bearing capacity factor which depends on the pile diameter and depth of embedment. The bearing capacity factor is typically taken as 9.

41

Figure 5-9: Chart for Estimating αt Coefficient and Bearing Capacity Factor N’q (Chart modified from Bowles, 1977)

42

800
Very Dense

Limiting Unit Toe Resistance, qL 400 (ksf)
200

600

Loose

Medium

Dense

Very Loose

0

30

35

40

45

Angle of Internal Friction, φ (degrees)

Figure 5-10: Relationship between maximum unit pile toe resistance and friction angle for cohesionless soils (after Meyerhof, 1976) STEP BY STEP PROCEDURE FOR α-METHOD IN COHESIVE SOIL (FHWA-HI-97-013) Step 1 Step 2 Delineate the soil profile into layers and determine the adhesion, ca, from Figure 5-11 or the adhesion factor, α, from Figure 5-12. For each soil layer, compute the unit shaft resistance, fs.

f s = c a = αc u
Step 3 Compute the shaft resistance in each soil layer and the ultimate shaft resistance, Rs, from the sum of the shaft resistances for each layer. Rs = ∑fsAs where: Step 4 As = pile shaft surface area = (perimeter) x (embedded length)

Compute the unit toe resistance, qt. qt = 9 c u

Step 5

Compute the ultimate toe resistance, Rt. Rt = qtAt

Step 6

Compute the ultimate pile capacity (kips) Qu = Rs + Rt

43

Step 7

Compute the allowable design load Qa (kips) Qa = Qu / Factor of Safety

2.0 1.6 1.2
Pile 0.8 Adhesion, ca (ksf)
D>40b

0.4 0.0 0.0

D=10b

1.0

2.0

3.0

4.0

Undrained Shear Strength, c u (ksf)
Timber Piles D = Distance from Ground Surface to Bottom of Clay Layer or Pile Toe; Whichever is Less b = Pile Diameter

Figure 5-11: Adhesion values for piles in cohesive soils (after Tomlinson, 1979) 5.5 EFFECTIVE STRESS METHOD FOR PILES IN COHESIONLESS AND COHESIVE SOILS

The long-term drained shear strength conditions of piles may be effectively modeled using effective stress methods. The effective stress method presented in this manual is based on the calculation of the unit shaft resistance (fs) using the following equation:

f s = βp o
where:

(5-8) = Bjerrum-Burland beta coefficient = Ks tanδ = Average effective overburden pressure along the pile shaft = Earth pressure coefficient = Friction angle between the pile and the soil

β po Ks δ

The unit toe resistance (qt) is calculated from:

q t = N t pt
where: Nt pt = Toe bearing capacity coefficient = Effective overburden pressure at the toe of the pile.

(5-9)

44

Figure 5-12: Adhesion factors for Driven Piles in Clay (after Tomlinson, 1980)

45

The toe bearing coefficient, Nt, and the beta coefficient, β, may be determined from Table 5-2 and Figures 5-13 and 5-14 may also be used to estimate the beta coefficient (β), and the toe bearing coefficient (Nt). Table 5-2 Range of β and Nt coefficients (Fellenius,1991) ϕ’ β 25 - 30 0.23 - 0.40 28 - 34 0.27 - 0.50 32 - 40 0.30 - 0.60 35 - 45 0.35 - 0.80

Soil Type Clay Silt Sand Gravel

Nt 3 - 30 20 - 40 30 - 150 60 - 300

1.0

Sand Silt

β

0.5

Clay

Gravel

Coefficient 0.4
0.3

0.2

Clay Silt Sand Gravel
20 25 30 35 40 45 50

φ (degrees)
Figure 5-13: Chart for Estimating β Coefficient versus Soil Type φ’ Angle (after Fellenius, 1991)

46

400 300 200 100

Toe Bearing Capacity Coefficient, Nt

50 40 30 20 10

5 4 3 2 20 25 30 35 40

Clay Silt Sand Gravel

45

50

φ (degrees)
Figure 5-14: Chart for Estimating Nt Coefficient versus Soil Type φ’ Angle (after Fellenius, 1991) STEP BY STEP PROCEDURE FOR EFFECTIVE STRESS METHOD (FHWA-HI-97-013) Step 1 Delineate the soil profile into layers and determine ϕ’ angle for each layer. A.) Construct the effective overburden versus depth diagram. B.) Divide the soil profile throughout the pile penetration depth into layers and determine the effective overburden pressure, po, at the midpoint of each layer. C.) Determine the ϕ’ angle for each layer from laboratory or in-situ test data. In the absence of laboratory or in-situ test data for cohesionless soils, determine the average corrected SPT N’ value for each layer and estimate ϕ’ angle from Table 5-1. Step 2 Select the β coefficient for each soil layer. Use Table 5-2 and Figure 5-13 to estimate β for each layer. Step 3 For each soil layer, compute the unit shaft resistance, fs.

f s = βp o

47

Step 4

Compute the shaft resistance in each layer of soil and the ultimate shaft resistance, Rs, from the sum of the shaft resistances from each layer. Rs = ∑fsAs where: As = pile shaft surface area = (perimeter) x (embedded length)

Step 5

Compute the unit toe resistance, qt.

q t = N t pt
Use local experience or Table 5-2 and Figure 5-14 to estimate Nt. Step 6 Compute the ultimate toe resistance, Rt. Rt = qtAt Step 7 Compute the ultimate pile capacity (kips) Qu = Rs + Rt Step 8 Compute the allowable design load Qa (kips) Qa = Qu / Factor of Safety 5.6 NOTTINGHAM AND SCHMERTMANN METHOD (Nottingham and Schmertmann, 1975) Static cone penetrometer test (CPT) data may be used when available to estimate the static capacity of single piles under axial loads. Nottingham and Schmertmann developed a procedure to estimate static pile capacity from CPT data. That procedure is summarized in the following paragraphs. The ultimate shaft resistance, Rs, in cohesionless soils may be derived from the unit sleeve friction of the CPT using the following equation:

R S = K 0. 5 fs As
where: K

[ (
fs

)

0 to 8 b

+ fs As

(

)

8 btoD

]

(5-10)

As b D

= Ratio of unit pile shaft resistance to unit cone sleeve friction from Figure 5-15 = Average unit sleeve friction over the depth interval indicated by the subscript (i.e., 0 to 8b) = Pile-soil surface area over fs depth interval = Pile diameter (average in depth interval) = Embedded pile length

48

If cone sleeve friction data is not available, Rs may be determined from the cone tip resistance in cohesionless soil as follows:

R S = C f ∑ q c AS
where: Cf qc AS = 0.018 for timber piles = Average cone tip resistance along the pile length = Pile-soil surface area

(5-11)

The shaft resistance in cohesive soils is obtained from the sleeve friction values using the following equation:

R S = α' fs AS
where:

(5-12) = Ratio of pile shaft resistance to cone sleeve friction Figure 5-16.

α’

Figure 5-17 is used to determine the ultimate pile toe capacity in cohesive soils using an average weighted cone resistance from 8 pile diameters above the toe to 3.75 pile diameters below the toe. The maximum value of qt should be limited to 100 ksf, unless local experience warrants use of higher values. STEP BY STEP PROCEDURE FOR NOTTINGHAM AND SCHMERTMANN METHOD (FHWAHI-97-013) Step 1 Delineate the soil profile into layers using the cone tip resistance, qt, and sleeve friction, fs , values. Compute the shaft resistance for each soil layer, RS. A.) For piles in cohesionless soils, compute ultimate shaft resistance, RS, using the average sleeve friction value for the layer, fs , and the K value. The K value should be determined using the full pile penetration depth to diameter ratio from Figure 7-18 and not the penetration depth of the layer. Conversely, the depth d corresponds to the pile toe depth, or the depth to the bottom of the layer, whichever is less.

Step 2

R S = K 0. 5 fs As

[ (

)

0 to 8 b

+ fs As

(

)

8 btoD

]

For cohesionless layers below a depth of 8b, the above equation for shaft resistance in a layer reduces to:

R S = K fs AS
For piles in cohesionless soils without sleeve friction data, compute the ultimate shaft resistance from:

R S = C f ∑ q c AS

49

B.) For piles in cohesive soils, compute the ultimate shaft resistance using the average sleeve friction value for the layer from:

RS = α' fs AS
Use Figure 5-16 to determine α′. Step 3 Step 4 Calculate the total pile shaft resistance from the sum of the shaft resistances from each soil layer. Compute the unit pile toe resistance, qt.

q t = (q c 1 + q c 2 ) ÷ 2
Use Figure 7-20 to determine qc1 and qc2. Step 5 Determine the ultimate toe resistance, Rt. Rt = qtAt Step 6 Compute the ultimate pile capacity (kips) Qu = Rs + Rt Step 7 Compute the allowable design load Qa (kips) Qa = Qu / Factor of Safety

50

K for Timber Piles
0 1.0 2.0 3.0 4.0

10

D/b
20 Electrical Penetrometer

30 Mechanical Penetrometer 40

Figure 5-15: Penetrometer design curves for pile side friction in sand (FHWA-TS-78-209)

1.4 1.2

Penetrometer to Pile 0.8 Friction Ratio, 0.6 α'
0.4 0.2 0.0 0.0 1.0 2.0 3.0 4.0 5.0 Timber Piles

1.0

Penetrometer Sleeve Friction, fs (ksf)

Figure 5-16: Design curve for pile side friction in clay (after Schmertmann, 1978)

51

Figure 5-17: Illustration of Nottingham and Schmertmann Procedure for Estimating Pile Toe Capacity (FHWA-TS-78-209)

52

5.7 UPLIFT CAPACITY OF SINGLE PILES The uplift capacity for timber piles in cohesive soils may be determined by considering the shaft resistance as presented in section 5.3 and adding the weight of the pile to obtain the ultimate uplift capacity. Comparison of uplift pile load tests with compression pile load tests in cohesive soils reveals that the uplift adhesion between the pile and the soil is approximately the same as the adhesion developed in compression. It has been found that negative pore pressures may occur in clays during uplift. The uplift capacity may, therefore, be less than the short-term capacity because the clay tends to soften with time as the negative pore pressure dissipates. For timber piles in cohesionless soils, the uplift capacity is generally less than the compression capacity of the pile. This lower capacity is a function of the taper of the pile and the skin friction between the pile and soil for uplift loading is less than for compression loading. FHWA, therefore, recommends that the design uplift capacity of a single pile in cohesionless or cohesive soils should be taken as one third (1/3) of the ultimate shaft resistance calculated from either the Nordlund method, the α method, the effective stress method, or the Nottingham and Schmertmann method. Two uplift connection details that are often used for timber piles are shown on Figure 5-18.

Figure 5-18: Uplift connection details.

53

Sign up to vote on this title
UsefulNot useful